Keynote: A Vision for Vision - Kubernetes in Its Second Decade - Tim Hockin

Don't miss out! Join us at our next Flagship Conference: KubeCon + CloudNativeCon Europe in Paris from March 19-22, 2024. Connect with our current graduated, incubating, and sandbox projects as the community gathers to further the education and advancement of cloud native computing. Learn more at kubecon.io
Keynote: A Vision for Vision - Kubernetes in Its Second Decade - Tim Hockin, Distinguished Engineer, Google
When Kubernetes started, we had a pretty good idea of what we wanted it to be. Almost 10 years later, we’re not too far off! Looking ahead, the problems we will have in the next 10 years will be VASTLY different from the previous 10 years. Nothing stands still. Everything has changed - our users, our contributors, and the industry we serve.
It’s easy for a project as big and broad as Kubernetes to fall into a pattern of refining the things we already have and missing out on the big picture. It's imperative that we pause and evaluate our trajectory and consider whether the things we’ve built over the last decade are still the things we’ll need for the next decade.
This talk will take a project-wide look at some of the decisions that got us where we are today, and will lay out some possible avenues for where we are going and what we think will be important. Maybe we’ll be right, and maybe we won’t - we’ll find out at KubeCon 2033!

[@alveedev]

love tim, great talk

[@DouglasRosser]

Kubernetes is pathologically eclectic. What else would let you choose your own storage, networking and container runtime interfaces? Would I change this? Maybe not, but K8S definitely needs more *idioms*. We can't be afraid of saying "Always choose X for your CRI unless your needs are Y". I know the CNCF doesn't want to pick the winners, but I'm afraid anything else will lead to madness.

[@vsoch]

I think it's reasonable that Kubernetes itself remains fairly agnostic about particular interfaces, drivers, or components. That landscape is constantly changing, and arguably a "best" or "always" is an ephemeral decision that is both context and community dependent. Groups that use Kubernetes can very easily derive suggestions of this type, and they have more freedom to change their mind.